Microfabricated lab-on-a-chip devices employing a fully integrated electrochemical (EC) detection system have been developed and evaluated. Both capillary electrophoresis (CE) channels and all CE/EC electrodes were incorporated directly onto glass substrates via traditional microfabrication techniques, including photolithographic patterning, wet chemical etching, DC sputtering, and thermal wafer bonding. Unlike analogous CE/EC devices previously reported, no external electrodes were required, and critical electrode characteristics, including size, shape, and placement on the microchip, were established absolutely by the photolithography process. For the model analytes dopamine and catechol, detection limits in the 4-5 microM range (approximately 200 amol injected) were obtained with the Pt EC electrodes employed here, and devices gave stable analytical performance over months of usage.

Miniaturized, battery-powered, high-voltage power supply, electrochemical (EC) detection, and interface circuits designed for microchip capillary electrophoresis (CE) are described. The dual source CE power supply provides +/- 1 kVDC at 380 microA and can operate continuously for 15 h without recharging. The amperometric EC detection circuit provides electrode potentials of +/-2 VDC and gains of 1, 10, and 100 nA/V. The CE power supply power is connected to the microchip through an interface circuit consisting of two miniature relays, diodes, and resistors. The microchip has equal length buffer and separation channels. This geometry allows the microchip to be controlled from only two reservoirs using fixed dc sources while providing a consistent and stable sample injection volume. The interface circuit also maintains the detection reservoir at ground potential and allows channel currents to be measured likewise. Data are recorded, and the circuits are controlled by a National Instruments signal interface card and software installed in a notebook computer. The combined size (4 in. x 6 in. x 1 in.) and weight (0.35 kg) of the circuits make them ideal for lab-on-a-chip applications. The circuits were tested electrically, by performing separations of dopamine and catechol EC and by laser-induced fluorescence visualization.

The aim of the study was to directly compare the threshold electrical charge density of the retina (retinal threshold) in rabbits for the generation of electrical evoked potentials (EEP) by delivering electrical stimulation with a custom-made microelectrode array (MEA) implanted into either the subretinal or suprachoroidal space. Nine eyes of seven Dutch-belted rabbits were studied. The electroretinogram (ERG), visual evoked potentials (VEP) and EEP were recorded. Electrodes for the VEP and EEP were placed on the dura mater overlying the visual cortex. The EEP was recorded following electrical stimulation of the MEA placed either subretinally beneath the visual streak of the retina or in the suprachoroidal space in the rabbit eye. An ab externo approach was used for placement of the MEA. Liquid perfluorodecaline (PFCL; 0.4 ml) was placed within the vitreous cavity to flatten the neurosensory retina on the MEA after subretinal implantation. The retinal threshold for generation of an EEP was determined for each MEA placement by three consecutive measurements consisting of 100 computer-averaged recordings. Animals were sacrificed at the conclusion of the experiment and the eyes were enucleated for histological examination. The retinal threshold to generate an EEP was 9 +/- 7 nC (0.023 +/- 0.016 mC cm(-2)) within the subretinal space and 150 +/- 122 nC (0.375 +/- 0.306 mC cm(-2)) within the suprachoroidal space. Histology showed disruption of the outer retina with subretinal but not suprachoroidal placement. The retinal threshold to elicit an EEP is significantly lower with subretinal placement of the MEA compared to suprachoroidal placement (P < 0.05). The retinal threshold charge density with a subretinal MEA is well below the published charge limit of 1 mC cm(-2), which is the level below which chronic stimulation of the retina is considered necessary to avoid tissue damage (Shannon 1992 IEEE Trans. Biomed. Eng. 39 424-6).

No abstract

A scalable and rather inexpensive solution to producing microanalytical systems with "on-chip" three-dimensional (3D) microelectrodes is presented in this study, along with applicability to practical electrochemical (EC) detection scenarios such as preconcentration and interferant removal. This technique to create high-aspect-ratio (as much as 4:1) gold microstructures in constrained areas involved the modification of stud bump geometry with microfabricated silicon molds via an optimized combination of temperature, pressure, and time. The microelectrodes that resulted consisted of an array of square pillars approximately 18 microm tall and 20 microm wide on each side, placed at the end of a microfabricated electrophoresis channel. This technique increased the active surface area of the microelectrodes by as much as a factor of 50, while mass transfer and, consequently, preconcentration collection efficiencies were increased to approximately 100%, compared to approximately 30% efficiency for planar nonmodified microelectrodes (samples that were used included the neurotransmitters dopamine and catechol). The 3D microelectrodes were used both in a stand-alone configuration, for direct EC detection of model catecholamine analytes, and, more interestingly, in dual electrode configurations for EC sample processing prior to detection downstream at a second planar electrode. In particular, the 3D electrodes were shown to be capable of performing coulometry or complete (100%) redox conversion of analyte species over a wide range of concentrations, from 4.3 microM to 4.4 mM, in either plug-flow or continuous-flow formats.